LGBMClassifier ajout des corrélations / prévisions dérivées

Zeus_LGBMRegressor.py

@@ -63,6 +63,7 @@ from sklearn.model_selection import GridSearchCV
 from sklearn.feature_selection import VarianceThreshold
 import seaborn as sns
 from lightgbm import LGBMRegressor
+import lightgbm as lgb
 
 # Couleurs ANSI de base
 RED = "\033[31m"
@@ -100,7 +101,8 @@ class Zeus_LGBMRegressor(IStrategy):
     #     'rsi_1h', 'rsi_deriv1_1h', 'rsi_deriv2_1h', "max_rsi_12_1h",
     # ]
 
-    model_indicators = ["ms-10", "ms-5", "ms-2", "ms-1", "ms-0"]
+    model = None
+    model_indicators = ["ms-10", "ms-5", "ms-4", "ms-3", "ms-2", "ms-1", "ms-0"]
     # model_indicators = ['open', 'high', 'close', 'haclose', 'percent', 'sma5', 'sma12', 'sma24', 'sma24_deriv1', 'sma24_deriv2', 'sma48', 'sma48_deriv1', 'sma48_deriv2', 'sma60', 'sma60_dist', 'sma60_deriv1',
     #    'sma60_deriv2', 'mid_smooth_3_deriv2', 'mid_smooth_12_deriv1', 'mid_smooth_12_deriv2', 'mid_smooth_24', 'mid_smooth_24_deriv1', 'mid_smooth_24_deriv2', 'max_rsi_12', 'max_rsi_24', 'max12',
     #    'max60', 'min60', 'min_max_60', 'bb_lowerband', 'bb_upperband', 'bb_width', 'macd', 'macdsignal', 'macdhist', 'sma_20', 'sma_100', 'atr', 'atr_norm', 'adx', 'obv', 'vol_24', 'adx_change',
@@ -993,12 +995,6 @@ class Zeus_LGBMRegressor(IStrategy):
                                             dataframe['mid_smooth_5h']
         dataframe["mid_smooth_5h_deriv2"] = 100 * dataframe["mid_smooth_5h_deriv1"].diff().rolling(window=60).mean()
 
-        dataframe["ms-10"] = dataframe["mid_smooth_24_deriv1"].shift(10)
-        dataframe["ms-5"] = dataframe["mid_smooth_24_deriv1"].shift(5)
-        dataframe["ms-2"] = dataframe["mid_smooth_24_deriv1"].shift(2)
-        dataframe["ms-1"] = dataframe["mid_smooth_24_deriv1"].shift(1)
-        dataframe["ms-0"] = dataframe["mid_smooth_24_deriv1"]
-        # dataframe["ms+10"] = dataframe["mid_smooth_24"].shift(-11)
         # ===============================
         # Lissage des valeurs Journalières
         horizon_d = 12 * 5 * 24
@@ -1114,6 +1110,14 @@ class Zeus_LGBMRegressor(IStrategy):
         #
         # self.model_indicators = usable_cols
 
+        dataframe["ms-10"] = dataframe["mid_smooth_24_deriv1"].shift(10)
+        dataframe["ms-5"] = dataframe["mid_smooth_24_deriv1"].shift(5)
+        dataframe["ms-4"] = dataframe["mid_smooth_24_deriv1"].shift(4)
+        dataframe["ms-3"] = dataframe["mid_smooth_24_deriv1"].shift(3)
+        dataframe["ms-2"] = dataframe["mid_smooth_24_deriv1"].shift(2)
+        dataframe["ms-1"] = dataframe["mid_smooth_24_deriv1"].shift(1)
+        dataframe["ms-0"] = dataframe["mid_smooth_24_deriv1"]
+        # dataframe["ms+10"] = dataframe["mid_smooth_24"].shift(-11)
         df = dataframe[self.model_indicators].copy()
 
         # Corrélations des colonnes
@@ -1572,7 +1576,7 @@ class Zeus_LGBMRegressor(IStrategy):
         # print(f"✅ Meilleur F1 : {f1s[best_idx]:.3f} au seuil {seuils[best_idx]:.2f}")
 
         # --- Après l'entraînement du modèle ---
-        preds = self.model.predict(X_test)
+        preds = model.predict(X_test)
 
         # --- Évaluation ---
         mse = mean_squared_error(y_test, preds)
@@ -1607,6 +1611,53 @@ class Zeus_LGBMRegressor(IStrategy):
 
         print(f"✅ Graphique sauvegardé : {plot_path}")
 
+        # save_dir = "/home/souti/freqtrade/user_data/plots/"
+        # os.makedirs(save_dir, exist_ok=True)
+
+        ax = lgb.plot_tree(model, tree_index=0, figsize=(30, 20),
+                           show_info=["split_gain", "internal_value", "internal_count"])
+        plt.title("Arbre de décision n°0")
+        plt.savefig(os.path.join(plot_dir, "lgbm_tree_0.png"), bbox_inches="tight")
+        plt.close()
+
+        ax = lgb.plot_tree(model, figsize=(40, 20))
+        plt.title("Vue globale du modèle LGBM")
+        plt.savefig(os.path.join(plot_dir, "lgbm_all_trees.png"), bbox_inches="tight")
+        plt.close()
+        # X_test = np.linspace(0, 10, 1000).reshape(-1, 1)
+        y_pred = model.predict(X_test)
+
+        plt.figure(figsize=(10, 5))
+        plt.plot(X_test, y_pred, color="red", label="modèle LGBM")
+        plt.title("Fonction apprise par LGBMRegressor")
+        plt.xlabel("x")
+        plt.ylabel("y")
+        plt.legend()
+        plt.savefig(os.path.join(plot_dir, "lgbm_function_curve.png"), bbox_inches="tight")
+        plt.close()
+
+        ax = lgb.plot_importance(model, max_num_features=30, figsize=(12, 6))
+        plt.title("Importance des features - LGBM")
+        plt.savefig(os.path.join(plot_dir, "lgbm_feature_importance.png"), bbox_inches="tight")
+        plt.close()
+
+
+        corr = X_train.corr() * 100  # en pourcentage
+
+        plt.figure(figsize=(20, 16))
+        sns.heatmap(corr, cmap="coolwarm", center=0, annot=False, fmt=".1f", cbar_kws={'label': 'Corrélation (%)'})
+        plt.title("Matrice de corrélation (%)")
+        plt.savefig(os.path.join(plot_dir, "correlation_matrix.png"), bbox_inches="tight")
+        plt.close()
+
+        plt.figure(figsize=(10, 6))
+        plt.scatter(y_test, model.predict(X_test), alpha=0.5)
+        plt.xlabel("Valeurs réelles")
+        plt.ylabel("Prédictions du modèle")
+        plt.title("Comparaison y_test vs y_pred")
+        plt.savefig(os.path.join(plot_dir, "ytest_vs_ypred.png"), bbox_inches="tight")
+        plt.close()
+
         print("\n===== ✅ FIN DE L’ANALYSE =====")
 
     def plot_threshold_analysis(self, y_true, y_proba, step=0.05, save_path=None):